- Rating
- 1
- Confidence
- 3
- Informativeness
- 2
- Presentation
- 1
- Soundness
- 3
Read review
Paper summary
This paper introduces FloydNet and Pivotal Attention, which extend the self-attention mechanism to second-order pair-state attention and further generalize it to higher orders.
Experiments on several expressiveness tests validate its structural discrimination capabilities. Further experiments on various graph-level, node-level, and link-level tasks show decent results.
Critical strengths
- The proposed method is sound and conceptually simple.
- Diverse empirical validation across multiple tasks.
Critical weaknesses
- Though already acknowledged by the authors, the complexity of -order attention and the number of states remain significant limitations that could restrict the proposed method to small-scale graphs.
- [Presentation] This paper is difficult to follow. For example, the term "relation" is used ambiguously. It is not until Sec. 2.2 that the authors define relations as "ordered node pairs", while terms like "target pair relation", "pair-relation", and "applied relational prediction" remain confusing. Meanwhile, are used without clear definitions, and is also used to denote the attention order.
- [Presentation] Figure 2 is potentially misleading. Figure 2A seems to imply that relation states are updated only for specific node pairs or (k)-order tuples. However, my understanding is that the states are updated for every node pair in the second-order case. Please also see the question below.
- [Presentation] It seems the authors never introduce these baselines, such as Subgraph GNN and Local 2-GNN.
- The paper seems to have a formatting issue, as it appears to use last year's template.
Questions for authors
- Is the target relation updated for every node pair ? If so, can the proposed method also be applied to non-graph data, such as sequential data?
- How is the proposed method applied to node-level tasks, given that the states are defined for node pairs ?